ChatGPT Revolutionizes Image Creation: Key Prompts and Tips

ChatGPT and Image Generation

ChatGPT, developed by OpenAI, now offers an innovative feature: the creation of original images from simple text instructions, known as prompts. This capability allows users to quickly generate visuals tailored to their needs, whether for creative concepts or communication materials. In just a few minutes, it is possible to produce ready-to-use visual elements, facilitating the exploration of concepts, the visual communication of ideas, and the adaptation of existing elements for different audiences, formats, or channels. Users can iterate rapidly by requesting variations, adjusting composition or size, or exploring new visual directions.

The Art of Writing Effective Prompts

To obtain quality images, it is essential to write clear and concise prompts. Generally, 1 to 3 sentences are sufficient to guide ChatGPT in creating an image. Key elements to include are the purpose of the image, the main subject, the ongoing action, the location, and the desired visual style. If specific details such as lighting or constraints need to be adhered to, it is important to mention them explicitly.

Clarity is more effective than clever phrasing, especially for details like layout, texture, materials, or lighting. For example, a precise description like "soft natural light coming from a window on the left" will be more effective than a vague formulation. When modifying existing images, it is crucial to be explicit about the desired changes, clearly indicating what should be altered or retained. A prompt such as "Change only X. Keep everything else exactly the same" is often the clearest way to guide a precise modification.

Examples and Practices for Successful Prompts

An example of a prompt could be: "Create a simple yet polished editorial illustration of a person learning a new skill in AI at their desk. Include a laptop, a notebook, books, sticky notes, and some subtle markers of progress like completed checkboxes, highlighted sections, or an organized plan pinned nearby. The person should look focused and engaged, with an overall calm, productive, and realistic atmosphere." For the style, use a clean and minimal background and a modern digital illustration style that feels approachable and neutral. Avoid logos and brand references, as well as sci-fi images or anything too abstract.

To improve results, it is advisable to proceed with small targeted revisions. Adjusting one element at a time, such as brightness or style, allows for maintaining consistency while refining the image. Specific instructions, such as "simplify the background" or "make the style more modern," are more effective than general feedback.

Advanced Tips for Optimal Images

When using multiple images, it is best to work with a small set to facilitate management. Each image should be clearly referenced, with instructions on how they relate to one another. For example: "Image 1 is a photo of my desk. Image 2 is a style reference. Apply the clean and minimal illustration style of Image 2 to Image 1 while keeping the same layout and objects."

To include text in an image, it is crucial to provide precise instructions regarding style, size, and placement. Text works best when instructions are very specific: put the text in quotes or in uppercase, specify the font style, size, color, and placement. Keep the text short, and for brand names or uncommon words, spell them out letter by letter.

Infographics and dense layouts require special attention to ensure clear text rendering. In some cases, it may be necessary to refine details in external design tools.

Ethical and Legal Considerations

When creating images of real people, it is important to use reference photos and obtain the necessary permissions. It is also advisable to favor generic designs to avoid any copyright violations. While attribution to OpenAI is not mandatory, it can be helpful to clarify the origin of the generated images.

Finally, it is essential to adhere to your organization's guidelines and OpenAI's usage policies when using these images.